Have you tried fetching the login page first?
I would suggest using Tamper Data to have a peek at exactly what's being sent when you request the login page and then log in normally using a web browser from a fresh start, with no initial cookies in place, so that your script can replicate it exactly.
That's the approach I used when writing the following, extracted from a script which needs to login to an Invision Power Board forum, using cookielib and urllib2 - you may find it useful as a reference.
import cookielib
import logging
import sys
import urllib
import urllib2
cookies = cookielib.LWPCookieJar()
opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cookies))
urllib2.install_opener(opener)
headers = {
'User-Agent': 'Mozilla/5.0 (Windows; U; Windows NT 5.0; en-GB; rv:1.8.1.12) Gecko/20080201 Firefox/2.0.0.12',
'Accept': 'text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5',
'Accept-Language': 'en-gb,en;q=0.5',
'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
}
# Fetch the login page to set initial cookies
urllib2.urlopen(urllib2.Request('http://www.rllmukforum.com/index.php?act=Login&CODE=00', None, headers))
# Login so we can access the Off Topic forum
login_headers = headers.copy()
login_headers.update({
'Referer': 'http://www.rllmukforum.com/index.php?act=Login&CODE=00',
'Content-Type': 'application/x-www-form-urlencoded',
})
html = urllib2.urlopen(urllib2.Request('http://www.rllmukforum.com/index.php?act=Login&CODE=01',
urllib.urlencode({
'referer': 'http://www.rllmukforum.com/index.php?',
'UserName': RLLMUK_USERNAME,
'PassWord': RLLMUK_PASSWORD,
}),
login_headers)).read()
if 'The following errors were found' in html:
logging.error('RLLMUK login failed')
logging.info(html)
sys.exit(1)